Yonatan Oren
San Francisco Bay Area
yonatano (at) cs.stanford.edu
github · linkedin

About

I'm building AI agents for offensive security at Armadin. I was previously a member of the research team at Together AI.

Experience

Together AI 2023–2025
Pre-training: RedPajama dataset. Inference & speculative decoding: blog post.
Stanford University
B.Sc. Mathematics (2020), M.Sc. Computer Science (2021–2023).

Papers

Proving Test Set Contamination in Black Box Language Models ICLR 2024 best paper runner-up
Oren*, Meister*, Chatterji*, Ladhak, Hashimoto
RedPajama: an Open Dataset for Training Large Language Models NeurIPS 2024 spotlight
Weber, Fu, Anthony, Oren, et al.
Distributionally Robust Language Modeling
Oren*, Sagawa*, Hashimoto*, Liang — EMNLP 2019
A Retrieve-and-Edit Framework for Predicting Structured Outputs
Hashimoto, Guu, Oren, Liang — NeurIPS 2018
Generating Sentences by Editing Prototypes
Guu*, Hashimoto*, Oren, Liang — TACL 2018